Create README.md
Browse filesModel Name: Perovskite-R1
Model Type: Domain-Specific Large Language Model (LLM)
Base Model: QwQ-32B
Overview
This model is a domain-specific large language model fine-tuned from QwQ-32B, specialized in the field of perovskite solar cells, particularly focusing on precursor additives. It is designed to assist researchers, engineers, and material scientists by providing knowledge, insights, and suggestions related to perovskite solar cell formulation, additive effects, and experimental design.
Intended Use
Research Support: Assist in understanding the role and mechanisms of precursor additives in perovskite solar cells.
Experimental Design: Provide guidance for selecting additives or predicting their impact on device performance.
Literature Summarization: Summarize and interpret scientific literature in the perovskite additive domain.
Hypothesis Generation: Suggest potential novel additive strategies for optimization.
Training Data
Fine-tuned on a curated dataset of academic papers and drug libraries related to perovskite solar cells and precursor additives.
Data includes information on additive types, concentrations, processing conditions, device performance metrics, and observed effects.
Model Architecture
Base: QwQ-32B
Fine-tuning: Instruction-tuning on domain-specific literature.
Capabilities: Answer questions, generate explanations, summarize research findings, and provide guidance on additive selection in perovskite solar cells.
Evaluation
Domain Accuracy: Tested on domain-specific literature QA tasks and additive effect prediction tasks.
Human-in-the-Loop Validation: Recommendations cross-checked with experimental reports to ensure relevance and accuracy.
Limitations: May hallucinate additive effects or interactions not reported in literature; not a replacement for experimental verification.
Reference
https://arxiv.org/abs/2507.16307